The reason why this was added is not valid anymore now that we set the QQC2 style in C++. Setting it causes crashes in QGuiApplications
that need to be worked around with an ugly hack in plasma-integration.
Differential Revision: https://phabricator.kde.org/D7998
Summary: As now it is set by our QPT
Reviewers: #plasma, gladhorn
Reviewed By: #plasma, gladhorn
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D7936
Summary:
Setting the style manually is necessary as there is only a variable that
applies to both quick controls versions, so it needs to be overwritten to not
cause a warning on every application start. The current way of setting it directly
got defeated by ugly bugs in QQuickControlSettings1, which need to be worked around.
See the added comments for an explanation.
Test Plan:
Set the variables manually, now works consistently everywhere without showing warnings.
For some reason I feel much dirtier now.
Reviewers: #plasma, mart
Reviewed By: #plasma, mart
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D7381
Summary:
It does explicitly what the comment above the export implies.
That way this warning on every use of Quick Controls 1 is gone:
WARNING: Cannot find style "org.kde.desktop" - fallback: "/usr/lib64/qt5/qml/QtQuick/Controls/Styles/Desktop"
Note that I'm not 100% sure that this does not have any
ill side effects as I don't quite understand what and where
org.kde.desktop is supposed to be.
Test Plan:
Started plasmashell and plasma-discover both with and without
the change, no difference except that the warning is now gone.
Reviewers: #plasma
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D7257
Summary:
Set the QtQuickControls style to our own: for QtQuickControls1
it will fall back to Desktop, while it will use our own org.kde.desktop
for QtQuickControlsStyle and Kirigami
Test Plan: systemsettings redesign uses qstyle qqc style, for both qqc1 and 2
Reviewers: #plasma, davidedmundson
Reviewed By: #plasma, davidedmundson
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D5720
Summary:
Remove outdated parts of startkde that are no longer used
Remove setion about kwin_failsafe which isn't used
Remove code related to oxygen fonts that aren't used
Remove section about darwin
Get rid of check that dbus-server exists, it autostarts
Don't try to set $qdbus to a path to qdbus... it wasn't used
conherently, so we don't really need it
Reviewers: #plasma, mart
Reviewed By: mart
Subscribers: mart, broulik, plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D2894
$qdbus is a full path to the qdbus binary, not relying on it being in
$PATH
It was completely unset in startplasma and startkde made a call to
ksplash using qdbus assuming it's in the path
Reviewed-by: David Edmundson
Summary: we can start launching ksmserver in the meantime
Test Plan: Logged in
Reviewers: #plasma, mart
Reviewed By: mart
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D2910
The check was supposed to check whether kinit would start and exit otherwise.
By adding the qdbus call inbetween, it would check for that one instead and if
no ksplash was running the exit code would be non zero aborting startup.
Differential Revision: https://phabricator.kde.org/D1888
Summary:
We want to drop workspace code from KDED as it doesn't make much sense
semantically.
This patch ignores the messages from kded, so that they can be dropped
in a future frameworks release and introduces a new signal sent from
startkde.
This puts all workspace related code in workspace.
Reviewers: #plasma
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D1859
Summary:
We use QT_SCREEN_SCALE_FACTORS as opposed to QT_SCALE_FACTOR as we need
to use one that will *NOT* scale logical DPI according to the scale.
Scaling the fonts makes sense if you don't also set a logical DPI
manually, but we need to do that for both PlasmaShell which does it's
own thing, and for KDE4/GTK2 applications.
Reviewers: #plasma
Subscribers: plasma-devel
Tags: #plasma
Differential Revision: https://phabricator.kde.org/D1849
Graphics seem more likely than anything else to be broken, if the user
explicitly selects failsafe mode from the DM we should follow the same
rationale as kwin and not trust the user's drivers.
REVIEW: 126205
Shutdown scripts are done by startkde after ksmserver quits. Which never
happens because we've told systemd to shutdown.
Old systems worked because they used to communicate with the display
manager which then closed us before shutting down, this is no longer the
case.
This moves handling the shutdown scripts into ksmserver (which already
handles startup scripts) and runs them before asking logind to shutdown.
BUG: 356190
REVIEW: 126268
The difference to ksyncdbusenv is that d-u-a-e can update systemd's
environment, too. This is only important if a userspace (dbus-daemon) bus
is used and service files make use of SystemdService to have dbus-daemon
tell systemd to start the service.
In the kernel bus (kdbus) case, systemd's environment gets updated when
org.freedesktop.dbus.UpdateActivationEnvironment is called.
REVIEW: 125768
1) it's not broken, that comment is ancient
2) the test is bust. DBus path can also be stored as an x property.
Before this patch the situation is worse as potentially we risk
launching two dbus daemons for the same session.
BUG: 352251
REVIEW: 125637
/bin/sh stops execution upon finding a nonzero exit code,
which means that startkde will not continue if a script in
.config/plasma-workspace/env/ produces an error.
This small fix will execute a no-op in such a case,
so that errors will not affect startkde's progress,
while still being visible in .xsession-errors.
REVIEW: 125323
BUG: 352491
On my system I faced the problem that instead of my prefix installed
ksmserver, the system installed ksmserver was started by kwrapper5.
REVIEW: 125113
Currently we are modifying the plugin path to include
$kdehome/lib/kde5/plugins, the original change says this is to allow
users to load styles and such.
In Plasma 5 $kdehome is .config, which shouldn't contain any plugins, so
this line currently isn't achieving anything.
Apparently instead it was causing problems and various distros are
already patching it out without issue.
BUG: 349371
FIXED-IN: 5.4.0
frameworkintegration is able to read fonts by itself, so this hardcode
makes it impossible for distributions to personalize fonts in globally
installed kdeglobal
If ~/.config does not exist readlink will not output anything,
causing startkde to write configs to /kstartupconfigkeys, which
is not a writable path for a regular user.
Since the original commit mentions that relative paths weren't working,
I've taken a alternative approach of using ${HOME} instead.
REVIEW: 119522
Currently env and shutdown scripts in startkde are not run due to the way kf5-config is used. There is no more $KDEHOME, so proper paths aren't being found.
This patch moves things to qtpaths, putting things under ConfigLocation + "plasma-workspace" + "env" (or "shutdown").
As a plus, kf5-config is no longer used in startkde.
CCBUG: 333793
REVIEW: 119145
KLOCALE_LANGUAGES was used by the kde4 ksplash in order to to know which
language to show. This environment variable is no longer used by the qml
based ksplash. It makes no sense to have it.
Additionally, this means we can stop linking against kdelibs4support.
This is important cause kdostartupconfig blocks the rest of the boot
sequence. On my system it causes a good 0.3 - 0.4 seconds delay. By no
longer linking to kdelibs4support it takes less than 0.1 seconds and no
longer shows up in the bootchat logs.
REVIEW: 119103